Health State Assessment Of Chain Tool Magazine Manipulator Based On State Information Fusion

In today's rapid development of the world,modern manufacturing as the cornerstone is playing an increasingly important role.Among them,machining centers have been widely used because of their advantages in multi-process automatic processing.The automatic tool change system is an indispensable functional part of the machining center.However,due to its low reliability and frequent failures,which have caused huge economic losses to the machine tool users,the health status of the automatic tool change system is accurately judged and appropriate The maintenance is particularly important.If you wait for the failure to occur,subsequent repairs will cause huge economic losses;if you repair in advance,there will be excess repairs,resulting in waste of production capacity.Therefore,how to accurately determine the current health status of the automatic tool change system and perform timely repairs and maintenance is a problem that needs to be solved urgently.This paper takes a certain type of chain tool magazine as the research object.First,the typical failure mode is explored through fault tree analysis.On this basis,a chain tool magazine health monitoring system is built;the integrated empirical mode decomposition is used to monitor the vibration.Signal feature extraction,and the use of particle swarm optimization algorithm to optimize the parameters of the support vector machine to construct the optimal support vector machine health monitoring model,through this model the health status of the chain tool magazine manipulator locking mechanism Evaluation.In addition,this paper uses the improved D-S evidence theory based on uncertainty analysis and analytic hierarchy process to fuse the data of each sensor to realize the health assessment of the chain tool magazine.The main research work of this paper is as follows:1.The working principle and failure analysis of the chain tool magazine and its manipulator.Introduce the parameters of the chain tool magazine,and explain the tool change process of the chain tool magazine.The fault tree analysis method is used to analyze the typical failure modes of the chain tool magazine to find out the root cause of the failure,and lay the foundation for the subsequent construction of the chain tool magazine health monitoring system.2.Build chain tool magazine test bench and condition monitoring system.On the basis of mastering the operation sequence and function of the chain tool magazine,the PLC control program of the tool magazine was compiled using Portal software.According to the cause of the typical failure of the tool magazine,select the vibration signal of the manipulator and the tool magazine,the current,voltage and temperature signals of the servo motor,the air pressure signal of the tool inverting cylinder,and the noise signal of the automatic tool change system as the state monitoring signal,using the Labview software A set of health monitoring system has been developed to realize signal collection,monitoring and storage.3.Judging the health status of the manipulator locking device through the vibration signal.Since the manipulator locking mechanism is the component with the highest failure rate in the chain tool magazine,and the vibration signal can better reflect the health of the tested component,this chapter uses the method of simulating the fatigue failure of the locking spring to judge the health of the manipulator locking device status.The vibration signal feature vector is extracted through ensemble empirical mode decomposition,and the parameters of the support vector machine are optimized by the particle swarm optimization algorithm.Finally,the optimized support vector machine is used to judge the health status of the manipulator locking device.4.Judge the overall health of the chain tool magazine.Select the manipulator Xdirection vibration,noise,motor current,motor temperature,and inverted tool cylinder air pressure as the health evaluation indicators of the chain tool magazine,and use the analytic hierarchy process and the uncertainty analysis method to monitor the signal of each sensor of the chain tool magazine.The importance and the amount of effective information are evaluated,the value of the generalized basic credibility is obtained through the parabolic membership function,and the DS evidence theory is finally used to integrate the five types of indicators to realize the health evaluation of the chain tool magazine.